{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

IVtestsTSLS

# IVtestsTSLS - 1 Estimating model with TSLS tsls CHILDREN 0...

This preview shows pages 1–3. Sign up to view the full content.

1. Estimating model with TSLS: tsls CHILDREN 0 EDUC AGE AGESQ; 0 FRSTHALF RADIO AGE AGESQ Model 1: TSLS, using observations 1-4361 (n = 4359) Missing or incomplete observations dropped: 2 Dependent variable: CHILDREN Instrumented: EDUC Instruments: const FRSTHALF RADIO AGE AGESQ coefficient std. error z p-value ---------------------------------------------------------- const -4.17235 0.302951 -13.77 3.74e-043 *** EDUC -0.0869102 0.0207104 -4.196 2.71e-05 *** AGE 0.332864 0.0166981 19.93 2.05e-088 *** AGESQ -0.00262977 0.000272898 -9.636 5.61e-022 *** Mean dependent var 2.267493 S.D. dependent var 2.221557 Sum squared resid 9283.501 S.E. of regression 1.460030 R-squared 0.568373 Adjusted R-squared 0.568076 F(3, 4355) 1839.741 P-value(F) 0.000000 Hausman test - Null hypothesis: OLS estimates are consistent Asymptotic test statistic: Chi-square(1) = 0.0326765 with p-value = 0.856551 Sargan over-identification test - Null hypothesis: all instruments are valid

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Test statistic: LM = 3.18698 with p-value = P(Chi-Square(1) > 3.18698) = 0.0742271 Weak instrument test - First-stage F-statistic (2, 4354) = 193.902 Critical values for desired TSLS maximal size, when running tests at a nominal 5% significance level: size 10% 15% 20% 25% value 19.93 11.59 8.75 7.25 Maximal size is probably less than 10% 2. Endogeneity test (Hausman test)
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}